0 PII: SO309-1740(98)00034-5

Meat Science, Vol. 49, No. 4, 447457, 1998 1998 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0309-1740/98 %19.00+0.00

Reference Methods for the Assessment of Physical Characteristics of Meat

Karl 0. Honikel*
Federal Centre for Meat Research, Kulmbach, Germany (Received 18 March 1998; accepted 22 March 1998)

ABSTRACT
As a spin-off of an OECD Workshop on pork quality, held in Helsinki in 1992, a group of scientists with many years of experience in theJield of meat quality assessment convened in February 1993 for the3rst time, and subsequently in 1994 and 1995, in Kulmbach at the German Federal Centre for Meat Research under the auspices of the OECD research project Management of Biological Resources. Three specific areas were discussed in order to develop internationally accepted reference methods: water-holding capacity tenderness 0 colour of meat.
l l

In the autumn of 1997 the methods were brought into theirjinalform at the Meat Industry Research Institute of New Zealand (MIRINZ). They are presented in this paper. 0 1998 Elsevier Science Ltd. All rights reserved

INTRODUCTION Despite many efforts over the years, there is still little consensus regarding methods of measuring the physical characteristics of meat and meat products. Many methods have been published in the scientific literature but only one procedure, to our knowledge, has been agreed upon internationally, and then only for beef (Boccard et al., 1981). Standardization of methods is essential if investigations carried out by different groups are to be directly comparable. Thus some agreement should be made regarding methods of measuring physical quality characteristics in meat and meat products. The lack of standard measurement is in contrast to accepted methods of measuring the chemical components of meat and meat products. In considering reference methodology, it was recognized that the techniques used to evaluate physical characteristics like water-holding capacity, tenderness and colour of meat could be applied for at least three different reasons:

*To whom correspondence

should be addressed. Fax: 0049 9221 803303.

441

448

K. 0. Honikel

1. as a quality assurance (QA) tool, within a processing operation 2. as an assessment of the effectiveness of production and processing treatments where there may be an interest in being able to compare results between laboratories or countries 3. as a research tool, in fundamental structural studies of muscle and meat. In the first case, a common methodology needs only to be appropriate for the plant or group of plants being controlled by specific QA programmes. The methods used should measure the desired characteristics necessary to monitor the process, but need not be comparable with other laboratories, where different criteria may be important. Where international comparison is important it is essential that methodologies be standardized. This would include all aspects of the testing procedure and this is area at which the reference methods are primarily directed. In contrast, where direct assessments are being made of the physical properties of meat as a function of structural (chemical or physical) changes, the experimental methodologies should not be constrained by reference methods. Instead researchers are encouraged to develop and use methodologies that enhance the precision and accuracy of testing methods and lead to an understanding of the basic mechanisms. It is likely that new understanding will lead, eventually, to methods that more closely predict consumer assessments of meat characteristics. GENERAL PRINCIPLES FOR ALL REFERENCE METHODS

The origin and husbandry of the live animal, slaughtering procedures and the post-mortem handling of the carcass should be described as precisely as possible. The description can include species, breed, sex, age, feeding regime, transport and preslaughter handling, slaughter conditions, chilling and ageing regime. The rate of pH and temperature decline post mortem, together with the final pH of the muscle should be reported. The history of the animal should preferably be known, although is it not always important. If it is known, it should be reported. REFERENCE
Introduction

METHODS

FOR THE ASSESSMENT CAPACITY IN MEAT

OF WATER-HOLDING

There is a multitude of procedures for measuring the water-hoding capacity (WHC) of meat and meat products (Burton-Gade et al., 1993). We have chosen to divide them according to the type of meat product and the process to which the meat has been subjected: 1. drip loss in raw, whole meat 2. water loss in cooked, whole meat. For each category, recommended methods are described together with their limitations.
Drip loss in raw whole meat
Principle

The mechanism of drip formation in raw, whole meat has been reviewed by Offer and Knight (1988). Water losses originate from volume changes of myofibrils induced by

Reference methods for the assessment of physical characteristics of meat

449

pre-rigor pH fall and the attachment of myosin heads to actin filaments at rigor where myofibrils shrink owing to pH fall. Denaturation of proteins may also contribute to a reduction in WHC particularly in conditions of rapid pre-rigor pH fall. The fluid thus expelled accumulates between fibre bundles. When a muscle is cut, this fluid will drain from the surface under gravity if the viscosity of the fluid is low enough and capillary forces do not retain it. This means that the methods chosen for measuring drip loss must conserve the integrity of the muscle before the sampling takes place to avoid external forces other than gravity. Orientation of the fibres with respect to cut is also important and should be taken into consideration. Surface evaporation has to be prevented and the method of supporting the meat piece should minimize tension (suspended from above) or compression (supported from below). For standardized meat samples, the following should be described: type of muscle, where on the muscle the sample is taken, muscle fibre orientation, surface area to weight ratio, time post mortem, temperature and pH.
Equipment

The equipment required is a balance of sufficient accuracy ( * 0.05 g), a sealable, waterimpermeable container (or plastic bag), sample support that allows the escape of fluid (plastic net bag or perforated support) and a temperature-controlled environment.
Procedure

Meat samples are cut from the carcass and immediately weighed. A sample weight of approximately 80-1OOg is recommended but other sample sizes may also be used. The samples are either placed in the netting and then suspended in an inflated bag, ensuring that the sample does not make contact with the bag, or placed within the container on the supporting mesh and sealed. After a storage period (usually 24 hr) at chill temperatures (1 to YC), samples are again weighed. The same samples can be used for further drip loss measurements, e.g. after 2, 7 days, etc., but in every case the initial weight is used as the reference point. At the time of measurement, samples should be taken immediately from the containers, gently blotted dry and weighed.
Calculation

Drip loss is expressed as a percentage of the initial weight.

Evaluation

At least two adjacent samples from the same muscle of similar weight and shape should be used. Triplicates are recommended.
Cooking loss in whole meat
Principle

During heating, the different meat proteins denature at varying temperatures (37-75C). Denaturation causes structural changes such as the destruction of cell membranes, transverse and longitudinal shrinkage of muscle fibres, the aggregation of sarcoplasmic proteins and shrinkage of the connective tissue. All these events, particularly the connective tissue changes, result in cooking losses in meat. Relevant reviews on the effect of heat on muscle proteins and structure have been given by Hamm (1977) and Offer (1984). Precautions taken regarding the geometry of the sample for the measurement of drip loss apply also to the cooking loss. Cooking conditions must be defined and controlled (heating rate and the end-point temperature at the thermal centre).

450 Equipment

K. 0. Honikel

The equipment required is a balance of sufficient accuracy ( * 0.05 g), a temperature-controlled water bath, thin-walled polyethylene bags and thermocouples to allow temperature recording in the centre of each sample.
Procedure

Samples should be freshly cut and weighed (initial weight). Individual standardized slices of 50 mm thick (maximum) and of a standard weight in thin-walled plastic bags are placed in a continously boiling water-bath, with the bag opening extending above the water surface. Samples should be cooked to a defined internal temperature; 75C is recommended. If other temperatures are used, these must be defined in the methodology. When the endpoint temperature has been attained, samples should be removed from the water-bath, cooled in an ice slurry and held in chill conditions (1 to SC) until equilibrated. The meat is then taken from the bag, blotted dry and weighed.
Calculation

The cooking loss is expressed as a percentage of the initial sample weight.

Evaluation

At least two samples of adjacent positions and similar weight and shape should be used. Triplicates are recommended.
Conclusion

It is essential that relevant factors that can affect the WHC values are defined as far as possible. These include the type of muscle and sample location within the muscle, meat quality parameters, such as the rate of pH decline, ultimate pH and details of the temperatures at which the samples or carcasses were maintained. The carcass chilling process (which affects chilling losses) is particularly important. REFERENCE
Introduction

METHODS

FOR THE ASSESSMENT

OF MEAT TENDERNESS

The methods and interpretation of meat tenderness measurements are highly variable. Although some attempts at standardization have taken place for instrumental (Boccard et al., 1981) and sensory techniques (Anon, 1978), they do not appear to have been universally accepted (Voisey, 1976). Tenderness measurement is important for both whole tissue and processed meats (Harris and Shorthose, 1988). However, the methodology discussed here has been restricted to whole tissue products since the nature of processed products and the requirements for objective measurement of their textural characteristics are diverse.
Field of application

The three methodologies described here will provide tenderness measurement, which can be related to sensory assessments. Each method has its advantages and limitations, but no single method provides a complete tenderness profile. All of the tests can be carried out in any of a wide variety of test frames, e.g. the Instron Universal Testing Instrument.

Reference method for the assessment of physical characteristics of meat

451

General principles

In describing the methods, we have started with the premise that measurement procedures must be well defined and accurately reported, regardless of which methodology is being used.
Samples

The muscle most widely used is the M. longissimus thoracis et lumborum. The sampling location must be clearly described (e.g. 1lth to 12th thoracic rib). Other muscles will also be tested and, when used, should be described with similar precision. It is recommended that, where possible, a slice of muscle perpendicular to the longitudinal axis of the muscle and thick enough to produce muscle fibre lengths of at least 50 mm along the fibre axis be used. This allows accurate preparation of test samples for all of the recommended test methods.
Storage of samples

Ideally, assessments should be performed immediately after sampling. If for convenience of testing the product is frozen, then conditions of freezing must be specified as these will affect the tenderness measurement. In such cases, samples should be packaged and frozen quickly and stored at -18C or below and the storage period should not exceed three months. Ideally cooking should be carried out from the frozen state. If thawing is necessary, then this must be specified as thawing will allow further ageing. The effects of freeze/thaw cycles on tenderness are variable (Locker and Daines, 1973) and may affect the results.
Cooking

The influence of cooking temperature and time on sample force deformation is large. At cooking temperatures up to 60C connective tissue influences predominate and above that myofibrillar components are more important. Therefore end-point temperatures of cooking in the centre need to be defined and measured accurately. Individual standardized slices of about 50mm thick and of a constant weight, in thinwalled plastic bags, are placed in a continoudy boiling water bath, with the bag opening extending above the water surface. Samples should be cooked to a defined internal temperature; 75C is recommended. If other temperatures are used these must be defined in the methodology. When the end-point temperature has been attained, samples should be removed from the waterbath, cooled in an ice slurry and then chilled until equilibrated. Under chilling conditions (l-SC), samples are stable for up to four days, when stored without cookout.
Testing

Objective tenderness measurements are intended to mimic the forces produced during biting and mastication. However, it is not always clear which structural properties of meat are described during meat tenderness evaluation although good correlations between subjective and objective evaluation are to be expected. Objective tenderness measurement can be used for raw or cooked meat; it is recommended that ten measurements per sample are made.
Tensile test method
Scope and principle

The tensile test is best suited for structural investigations (Purslow, 1985) rather than to predict sensory evaluation of tenderness. It is a useful test in conjunction with other methods. The test can be carried out on raw or cooked meat but if it is conducted on cooked meat, the cooking procedure should be as specified earlier. Results will be affected

452

K. 0. Honikel

by sample size and by strain rate, but this latter effect is small. Problems with gripping the samples are a major cause of measurement failure, especially with raw meat. Cyanacrolate adhesives or freezing grips can be employed (Lewis and Purslow, 1991).
Procedure

The block of cooked (or raw) meat should be sliced with a thin-bladed sharp knife to minimize damage. The standard thickness of the slices is 3.5 mm but for some species and some muscles thinner slices may be required. Testing may be conducted either transverse or parallel to the fibre direction. The standard slices of the tensile test samples will be cut using a template (Fig. 1) to define dimension and shape. If smaller samples are required because of physical restrictions imposed by muscle size and shape, then the proportions of 4:1:0.5 for 1ength:width:thickness should be maintained. When cutting the samples to the dumb-bell shape, a continuous cut to produce a smoothly contoured surface is required. Great care should be taken to ensure that fibre direction remains in the required parallel or transverse orientation throughout the sample. Dumb-belling is less important for tensile tests transverse to the fibre direction where parallel-sided strips may be used provided that (a) fracture occurs away from the edges of the grips and (b) a length between grips to width ratio of 4:l is maintained. After cutting, the width and thickness of the samples should be measured with vernier calipers, taking care not to damage the sample. When the degree of variation is established it may not be necessary to measure every sample. However, it must be recognized that the cross-sectional area of the sample will affect the results. Specimens should be subjected to extension at a strain rate of, for example, 2 per minute (i.e. strain rate=extension rate/specimen length). Thus, for the suggested 28mm length, an extension rate of 56 mm/min is recommended. A rate of 50 mm/min would be acceptable on test machines with limited preset speeds. The sample will normally be gripped with pneumatic clamps at pressures sufficient to maintain a firm grip without obvious slippage, yet minimizing specimen damage.
Calculation

A load deformation curve to complete rupture should be obtained. Criteria for the acceptance of test results is that fracture should occur in the parallel-sided region of the specimen. Breaking stress is defined as: breaking stress = peak force/measured width x thickness The results should be given in pascals (1 Pa is equivalent to 1 N/m*). Other parameters can be measured, including the total energy to fracture (area under the curve) and breaking strain (breaking strain = extension of peak force/original gauge length). Because of variability, a minimum of eight to ten specimens should be tested. Larger amounts of connective tissue in the samples tested cause high variability.
28mm

Fig. 1. Shape of template

for tensile test.

Reference methods for the assessment of physical characteristics of meat

453

Warner Bratzler shear test

Equipment

Although Warner Bratzler devices and sample configuration are extremely variable, the recommended equipment is as follows. The blade should be 1.2mm thick with a rectangular hole 11 mm wide and at least 15 mm high. The hole should have square smooth edges and the blade should be drawn or be pushed at 5&100mm/min between side plates positioned to provide a minimum gap between blade and plates. A means of holding the sample may be required with some configurations.
Procedure

The sample should be cut from a block of cooked meat and taken to avoid damage. Sample strips should be cut with a lOOmm* (10x 10) cross-section with the fibre direction parallel to a long dimension of at least 30 mm. The sample should be sheared at right angles to the fibre axis. Units of measurement are kPa. Often the simplified units are kgf or N.
Evaluation

The parameters to be measured from the force deformation curve (Fig. 2) are the peak force (the maximum recorded) and the total energy. Initial yield may be useful in some instances but will not always be apparent. A minimum of eight to ten samples should be tested.

Penetrometer measurements
Equipment

A cylindrical
Procedure

flat ended plunger

(diameter

1.13 cm, area 1 cm*) in a test frame.

The plunger is driven (lOOmm/min) vertically 80% of the way through a 1 cm thick meat sample cut so that the fibre axis is perpendicular to the direction of the plunger penetration. The plunger is driven twice into the meat at each location.

PFIY

Initial Yield

Distance
Fig. 2. Force deformation curve of the Warner Bratzler shear force measurement.

454

K. 0. Honikel

1st down stroke

1st up stroke

2nd down stroke

2nd up stroke

Plunger position
Fig. 3. Work and force deformation curve of penetrometer

measurements.

Evaluation The work and force-deformation hardness: cohesiveness: gumminess: curves are recorded (Fig. 3):

maximal force (N) for first deformation (H2) ratio of work done during the second penetration, (AZ/Al) hardnessxcohesiveness. (see Fig. 3). A minimum

relative

to the first

Other parameters can also be defined should be measured.

of eight to ten samples

Conclusion
When testing is used to evaluate consumer products it is strongly recommended that the methods should be validated against sensory panels. The reference methods are advanced as appropriate at this time but it is stressed that development of new techniques is likely as researchers explore the mechanical properties of meat and the effects of different sample preparation techniques. The ideal of a single measurement to accurately predict consumer perceptions under all conditions may, as yet, not be achievable. REFERENCE METHOD FOR MEAT COLOUR MEASUREMENT

Introduction
Colour is the visual characteristic of meat that gives the critical first impression and can be measured both subjectively and instrumentally. Guidelines for human evaluation of meat colour have been published by AMSA (1991).

Reference methods for the assessment of physical characteristics

of meat

455

Field of application

There are three sources of colour variation in meat: 1. the content of pigment (myoglobin) is intrinsic to the muscle, being dependent on primary production factors such as species breed, age of animal and nutritional status; 2. the preslaughter period, the slaughter process and subsequent processing affect colour by influencing the rate and extent of pH and temperature decline 3. during storage, distribution and display, the processes of oxygenation and oxidation of myoglobin influence colour.
Method

History and specljication of the meat Sampling should be conducted after the ultimate pH is reached. The muscle must be clearly specified and the location within the muscle described. In general, sampling should be a cross-section taken perpendicular to the long axis of the muscle, and the sample should have a minimum thickness of 1.5cm or, preferably, 2cm. In the case of meat with very low myoglobin levels, the relationship between sample thickness and light transmittance can be checked by measuring against both white and black backgrounds. If meat is stored prior to exposing a surface for colour measurement, it should be refrigerated at no higher than 3C. Storage conditions such as temperature, exposure to light and packaging must be specified. Blooming The time required for blooming, when the surface myoglobin is fully oxygenated, is important and is dependent on such factors as species, temperature of sample and time after rigor. It is recommended that blooming be allowed for at least 1 hr or, preferably, 2 hr (the duration of blooming must be defined) at a maximum temperature of 3C. Surface drying must be avoided by use of an oxygen-permeable but water-impermeable film, or by control of humidity. Subsequent colour measurement may be made with or without the film in place, depending on the instrumentation. Testing It is recommended that at least triplicate measurements be made on different sites of the exposed surface. It must be recognized that in some species and muscles there are considerable colour differences between lateral and medial sites on the muscle cross-section. Muscles with high levels of intramuscular fat (marbling), or collagen, are likely to produce highly variable readings. Equally, discolouration of meat surfaces can cause sampling difficulties and these should be reported. Procedure The recommended parameters are a light source of D 65 with the illumination/viewing system as 45/O or O/45 or diffuse/8 (d/8). The recommended standard observer angle is 10 and the color scale is the L*a*b* (Commission International de lEclairage, 1976). Calibration The aperture should be as large as possible (within the limitations of the sample to be measured). The instrument must be warmed according to the manufacturers instructions. Specular reflectance should be excluded if this is within the capacities of the instrument. The calibration should be based upon a black standard as L* = 0 and white standard

456

L 0. Honikel

(equivalent to BaS04 or freshly burnt MgO) as L* = 100. If sample measurements made through an overwrap film, the calibration must be similarly covered.
Evaluation

are

If other parameters are used for colour measurement then they must be specified in the method. It is the experience of the expert group that even when the recommended parameters are used, different results can be obtained by different instruments within the same laboratory. This may be because of differences in instrumental design such as aperture size, halogen versus xenon lamp, illumination/viewing system, 45/O versus diffuse/8 (d/8). Variability also occurs because some instruments measure colour from an area that is less than the illuminated area, which minimizes edge effects because of translucency. It is recommended that a meat-like spectral reflectance standard should be developed to improve comparisons between published results. The recommendations made herein are for the purpose of standardizing the method for the measurement of fresh meat colour. It is recognized that determining colour stability is another important criterion in fresh meat, but this may require alternative measurement procedures.

ACKNOWLEDGEMENTS The following scientists have contributed to the reference methods for water-holding capacity reference methods: Barton-Gade, P. A., Danish Meat Research Institute, Roskilde, Denmark. Chrystall, B. B., Daly, C. C., Simmons, N. J., Meat Industry Research Institute, Hamilton, New Zealand. Demeyer, D., De Smet, S., Department of Animal Production, University of Ghent, Melle, Belgium. Honikel, K.O., Federal Centre -for Meat Research, Kulmbach, Germany. Joseph, R. L., The National Food Centre, Dublin, Ireland. Puolanne, E., Department of Food Technology, University of Helsinki, Finland. Severini, M., Institute for Science and Hygiene of Food of Animal Origin, Faculty of Veterinary Medicine, Perugia, Italy. Smulders, F. J. M., Veterinarmedizinische Universitat Wien, Institut fur Fleischhygiene, Fleischtechnologie u. Lebensmittelkunde, Wien, Osterreich. Tornberg, E., Swedish Meat Research Institute, Kavlinge, Sweden. To the reference methods for the assessment of meat tenderness contributed: Chrystall, B. B., Daly, C. C., Gilbert, K. V., Simmons, N. J. and Young, O., A. Meat Industry Research Institute, Hamilton, New Zealand. Culioli, J., Meat Research Laboratory, INRA, Theix, StGenes-Champanelle, France. Demeyer, D. and Uytterhaegen, L., Department of Animal Production, University of Ghent, Melle, Belgium. Honikel, K.O. and Schwagele, F., Federal Centre for Meat Research, Institute for Chemistry and Physics, Kulmbach, Germany. Moller, A.J. and Purslow, P., Department of Dairy and Food Science, The Royal Veterinary and Agricultural University, Frederiksberg-Copenhagen, Denmark. Shorthose, W.R., Meat Research Laboratory, CSIRO, Cannon Hill, Queensland, Australia. To the reference methods for the assessment of meat colours contributed: Chrystall, B. B., Daly, C. C., Gilbert, K. V., McNichol, S., Simmons, N. J., Young, 0. A., Meat Industry Research, Hamilton, New Zealand. Cassens, R.G., University of Wisconsin, Madison, WI, USA. Demeyer, D., De Smet, S., University of Ghent, Melle, Belgium. Eikelenboom, G., Institute for Animal Science and Health, Zeist, The Netherlands. Honikel, K. O., Bundesanstalt ftir Fleischforschung, Kulmbach, Germany. Johansson, G., Swedish Meat Research Institute, Kaviinge, Sweden. Nielsen, T., Danish Meat Research Institute, Roskilde, Denmark. Renerre, M., INRA Theix, France. Richardson, I., University of Bristol, Langford, Bristol, UK. Sakata, R., Azabu University, Fuchinobe, Japan.

Reference methods for the assessment

of physical characteristics

of meat

457

All contributing authors and especially myself as the initiator and coordinator for the reference methods wish to thank the OECD in Paris for the financial support needed to produce this document.

REFERENCES
(1978) Guidelines for the Cookery and Sensory Evaluation of Meat. American Meat Science Association and National Livestock and Meat Board, Chicago. AMSA (1991) Guidelines for Meat Color Evaluation. American Meat Science Association and National Livestock and Meat Board, Chicago, IL. Barton-Gade, P. A., Demeyer, D., Honikel, K. O., Joseph, R. L., Puolanne, E., Severini, M., Smulders, F. and Tornberg, E. (1993) Final Version (I) of Reference Methods for Water Holding Capacity in Meat and Meat Products. Procedures Recommended by an OECD Working Group and Prepresented at the 39th ICoMST in 1993. In: Proceedings 39th Intern. Congress of Meat Science and Technology, Calgary, Canada, file S4, PO2.WP. Boccard, R., Buchter, L., Casteels, E., Cosentino, E., Drandsfield, E., Hood, D. E., Joseph, R. L., MacDougal, D. B., Rhodes, D. N., Schon, I., Tinbergen, B. J. and Touraille, C. (1981) Procedures for measuring meat quality characteristics in beef production experiments. Report of a working group in the Commission of the European Communities (CEC) Beef Production Research Programme. Livest. Prod. Sci. 8, 385-397. Commission International de 1Eclairage (1976) Colorimetry, 2nd edn. CIE, Vienna. Hamm, R. (1977) In Physical, Chemical and Biological Changes in Food caused by Thermal Processing, eds T. Hoyem and 0. Kvale, p. 101. Applied Science Publishers, London. Harris, P. V. and Shorthose, W. R. (1988) Meat texture. In Developments in Meat Science 4, ed. R. Lawrie, Elsevier Applied Science, London, p. 2455296. Lewis, G. J. and Purslow, P. P. (1991) The effect of marination and cooking on the mechanical properties of intramuscular connective tissue. Journal of Muscle Foods 2, 177-195. Locker, R. H. and Daines, G. J. (1973) The effect of repeated freeze-thaw cycles on tenderness and cooking loss in beef. J. Sci. Fd. Agric. 24, 1273-1275. Offer, G. (1984) Progress in the biochemistry, physiology and structure of meat. In Proceedings of the 30th European Meeting of Meat Research Workers, p. 87-94. Bristol, U.K. Offer, G. and Knight, P. (1988) The structural basis of water-holding in meat; Part 2: Drip Losses. In Developments in Meat Science 4, ed. R. Lawrie, p. 1733241. Elsevier, Oxford. Purslow, P. P. (1985) The physical basis of meat texture: observations on the fracture behaviour of cooked bovine m. semitendinosus. Meat Science 12, 39-60. Voisey, P. (1976) Engineering assessment and critique of instruments used for meat tenderness evaluation. Journal of Texture Studies 7, 1148. Anon